Re: the future of xslt

On Sat, Jun 21, 2008 at 05:08:15PM -0400, G. Ken Holman wrote:
[...]
> Or perhaps XML is being used more these days for element content 
> rather than mixed content and the "pull model" of XQuery satisfies 
> those who do not need the "push" model of XSLT.  Learning XSLT covers 
> off both pull and push, so my money is still on XSLT as the 
> technology to learn ... though I'm sure you won't hear that from 
> XQuery vendors.

At least some of them also support XSLT.

> For publishing solutions with XSL-FO my customers invariably are 
> using mixed content, so XQuery doesn't even come into the 
> equation.

I'm worried this will confuse people.  XQuery does of course
handle mixed content data, and shares the same data model
as XPath 2.0 and XSLT 2.0, too.  It's true that it does not
have apply-templates; XQuery is best when you are grabbing
fragments of documents; if you want to transform the fragments,
it can also make sense to use XSLT after XQuery.

[...]
> ... all because the XSLT files are themselves XML and suitable for 
> XSLT transformation into the Javadoc-like documentation.
[...]
> Given that XQuery documents are not XML documents, this wouldn't be 
> possible.

Actually XQueryX documents are even more amenable to machine
processing than XSLT in some ways.

But it's not one or the other, the technologies are not completing,
but complementary.

Best,

Liam

-- 
Liam Quin, W3C XML Activity Lead, http://www.w3.org/People/Quin/
http://www.holoweb.net/~liam/ * http://www.fromoldbooks.org/